Natural language processing to automate a web-based model of care and modernize skin cancer multidisciplinary team meetings.

BACKGROUND: Cancer multidisciplinary team (MDT) meetings are under intense pressure to reform given the rapidly rising incidence of cancer and national mandates for protocolized streaming of cases. The aim of this study was to validate a natural language processing (NLP)-based web platform to automate evidence-based MDT decisions for skin cancer with basal cell carcinoma as a use case. METHODS: A novel and validated NLP information extraction model was used to extract perioperative tumour and surgical factors from histopathology reports. A web application with a bespoke application programming interface used data from this model to provide an automated clinical decision support system, mapped to national guidelines and generating a patient letter to communicate ongoing management. Performance was assessed against retrospectively derived recommendations by two independent and blinded expert clinicians. RESULTS: There were 893 patients (1045 lesions) used to internally validate the model. High accuracy was observed when compared against human predictions, with an overall value of 0.92. Across all classifiers the virtual skin MDT was highly specific (0.96), while sensitivity was lower (0.72). CONCLUSION: This study demonstrates the feasibility of a fully automated, virtual, web-based service model to host the skin MDT with good system performance. This platform could be used to support clinical decision-making during MDTs as 'human in the loop' approach to aid protocolized streaming. Future prospective studies are needed to validate the model in tumour types where guidelines are more complex.

Epilepsy genetics: a practical guide for adult neurologists.

An understanding of epilepsy genetics is important for adult neurologists, as making a genetic diagnosis gives clinical benefit. In this review, we describe the key features of different groups of genetic epilepsies. We describe the common available genetic tests for epilepsy, and how to interpret them.

Longitudinal analysis of the relationship between motor and psychiatric symptoms in idiopathic dystonia.

BACKGROUND AND PURPOSE: Although psychiatric diagnoses are recognized in idiopathic dystonia, no previous studies have examined the temporal relationship between idiopathic dystonia and psychiatric diagnoses at scale. Here, we determine rates of psychiatric diagnoses and psychiatric medication prescription in those diagnosed with idiopathic dystsuponia compared to matched controls. METHODS: A longitudinal population-based cohort study using anonymized electronic health care data in Wales (UK) was conducted to identify individuals with idiopathic dystonia and comorbid psychiatric diagnoses/prescriptions between 1 January 1994 and 31 December 2017. Psychiatric diagnoses/prescriptions were identified from primary and secondary health care records. RESULTS: Individuals with idiopathic dystonia (n = 52,589) had higher rates of psychiatric diagnosis and psychiatric medication prescription when compared to controls (n = 216,754, 43% vs. 31%, p < 0.001; 45% vs. 37.9%, p < 0.001, respectively), with depression and anxiety being most common (cases: 31% and 28%). Psychiatric diagnoses predominantly predated dystonia diagnosis, particularly in the 12 months prior to diagnosis (incidence rate ratio [IRR] = 1.98, 95% confidence interval [CI] = 1.9-2.1), with an IRR of 12.4 (95% CI = 11.8-13.1) for anxiety disorders. There was, however, an elevated rate of most psychiatric diagnoses throughout the study period, including the 12 months after dystonia diagnosis (IRR = 1.96, 95% CI = 1.85-2.07). CONCLUSIONS: This study suggests a bidirectional relationship between psychiatric disorders and dystonia, particularly with mood disorders. Psychiatric and motor symptoms in dystonia may have common aetiological mechanisms, with psychiatric disorders potentially forming prodromal symptoms of idiopathic dystonia.

Epilepsy, antiepileptic drugs, and the risk of major cardiovascular events.

OBJECTIVE: This study was undertaken to determine whether epilepsy and antiepileptic drugs (including enzyme-inducing and non-enzyme-inducing drugs) are associated with major cardiovascular events using population-level, routinely collected data. METHODS: Using anonymized, routinely collected, health care data in Wales, UK, we performed a retrospective matched cohort study (2003-2017) of adults with epilepsy prescribed an antiepileptic drug. Controls were matched with replacement on age, gender, deprivation quintile, and year of entry into the study. Participants were followed to the end of the study for the occurrence of a major cardiovascular event, and survival models were constructed to compare the time to a major cardiovascular event (cardiac arrest, myocardial infarction, stroke, ischemic heart disease, clinically significant arrhythmia, thromboembolism, onset of heart failure, or a cardiovascular death) for individuals in the case group versus the control group. RESULTS: There were 10 241 cases (mean age = 49.6 years, 52.2% male, mean follow-up = 6.1 years) matched to 35 145 controls. A total of 3180 (31.1%) cases received enzyme-inducing antiepileptic drugs, and 7061 (68.9%) received non-enzyme-inducing antiepileptic drugs. Cases had an increased risk of experiencing a major cardiovascular event compared to controls (adjusted hazard ratio = 1.58, 95% confidence interval [CI] = 1.51-1.63, p < .001). There was no notable difference in major cardiovascular events between those treated with enzyme-inducing antiepileptic drugs and those treated with non-enzyme-inducing antiepileptic drugs (adjusted hazard ratio = .95, 95% CI = .86-1.05, p = .300). SIGNIFICANCE: Individuals with epilepsy prescribed antiepileptic drugs are at an increased risk of major cardiovascular events compared with population controls. Being prescribed an enzyme-inducing antiepileptic drug is not associated with a greater risk of a major cardiovascular event compared to treatment with other antiepileptic drugs. Our data emphasize the importance of cardiovascular risk management in the clinical care of people with epilepsy.

SUDEP and mortality in epilepsy: The role of routinely collected healthcare data, registries, and health inequalities.

Routinely collected data are a powerful research resource and offer the opportunity to further our understanding of epilepsy mortality and Sudden Unexpected Death in Epilepsy (SUDEP). The advantages of using routinely collected data include that it often covers whole populations, is already collected, and can be easily linked to other data sources. A significant disadvantage is the difficulty in obtaining accurate causes of death and correctly identifying cases of SUDEP. Using and linking data from epilepsy death registries can improve the quality of mortality data for research. Epilepsy prevalence, incidence, and mortality rates are associated with socioeconomic deprivation. Further research into understanding the link between deprivation and epilepsy mortality could lead to ways to reduce epilepsy mortality. This paper is for the Special Issue :Prevent 21: SUDEP Summit - Time to Listen.

Treatment of psychiatric comorbidities in patients with epilepsy and intellectual disabilities: Is there a role for the neurologist?

This paper will explore the nature of psychiatric co-morbidities in people with an intellectual disability (ID) who have epilepsy. The complexity of clinical presentations and associated co-morbidities require thorough assessment utilising both neurological and psychiatric skills. The neurologist plays a central role in the management of epilepsy in people with ID and therefore requires basic competencies in the assessment of neuropsychiatric co-morbidities. This is key to liaison with other specialist services to ensure individuals receive holistic person-centred care. This article is part of the Special Issue "Obstacles of Treatment of Psychiatric Comorbidities in Epilepsy".

Educational attainment of children born to mothers with epilepsy.

OBJECTIVE: Small prospective studies have identified that children exposed to valproate in utero have poorer scores on cognitive testing. We wanted to identify whether children exposed to antiepileptic drugs (AEDs) in utero have poorer school performance. METHODS: We used anonymised, linked, routinely collected healthcare records to identify children born to mothers with epilepsy. We linked these children to their national attainment Key Stage 1 (KS1) tests in mathematics, language and science at the age of 7 and compared them with matched children born to mothers without epilepsy, and with the national KS1 results. We used the core subject indicator (CSI) as an outcome measure (the proportion of children achieving a minimum standard in all subjects) and the results in individual subjects. RESULTS: We identified 440 children born to mothers with epilepsy with available KS1 results. Compared with a matched control group, fewer children with mothers being prescribed sodium valproate during pregnancy achieved the national minimum standard in CSI (-12.7% less than the control group), mathematics (-12.1%), language (-10.4%) and in science (-12.2%). Even fewer children with mothers being prescribed multiple AEDs during pregnancy achieved a national minimum standard: CSI (by -20.7% less than the control group), mathematics (-21.9%), language (-19.3%) and science (-19.4%). We did not observe any significant difference in children whose mothers were prescribed carbamazepine or were not taking an AED when compared with the control group. CONCLUSIONS: In utero exposure to AEDs in combination, or sodium valproate alone, is associated with a significant decrease in attainment in national educational tests for 7-year-old children compared with both a matched control group and the all-Wales national average. These results give further support to the cognitive and developmental effects of in utero exposure to sodium valproate as well as multiple AEDs, which should be balanced against the need for effective seizure control for women during pregnancy.

Revisiting basal cell carcinoma clinical margins: Leveraging natural language processing and multivariate analysis with updated Royal College of Pathologists histological reporting standards.

INTRODUCTION: Data supporting the current British Association of Dermatologists guidelines for the management of basal cell carcinoma (BCC) are based on historic studies and do not consider the updated Royal College of Pathologists (RCPath) histological reporting standards. The aim of this study was to use natural language processing (NLP)-derived data and undertake a multivariate analysis with updated RCPath standards, providing a contemporary update on the excision margins required to achieve histological clearance in BCC. METHODS: A validated NLP information extraction model was used to perform a rapid multi-centre, pan-specialty, consecutive retrospective analysis of BCCs, managed with surgical excision using a pre-determined clinical margin, over a 17-year period (2004-2021) at Swansea Bay University Health Board. Logistic regression assessed the relationship between the peripheral and deep margins and histological clearance. RESULTS: We ran our NLP algorithm on 34,955 BCCs. Out of the 1447 BCCs that met the inclusion criteria, the peripheral margin clearance was not influenced by the BCC risk level (p = 0.670). A clinical peripheral margin of 6 mm achieved a 95% histological clearance rate (95% confidence interval [CI], 0.93-0.98). Tumour thickness inversely affected deep-margin histological clearance (OR 0.720, 95% CI, 0.525-0.991, p < 0.05). Depth level 2 had a 97% probability of achieving deep-margin histological clearance across all tumour thicknesses. CONCLUSION: Updated RCPath reporting standards minimally impact the peripheral margin histological clearance in BCC. Larger clinical peripheral margins than those indicated by current guidelines may be necessary to achieve excision rates of ≥95%. These findings emphasise the need for continuous reassessment of clinical standards to enhance patient care.

Assessing the extent and determinants of socioeconomic inequalities in epilepsy in the UK: a systematic review and meta-analysis of evidence.

BACKGROUND: Socioeconomic inequalities in epilepsy incidence and its adverse outcomes are documented internationally, yet the extent of inequalities and factors influencing the association can differ between countries. A UK public health response to epilepsy, which prevents epilepsy without widening inequalities, is required. However, the data on UK epilepsy inequalities have not been synthesised in a review and the underlying determinants are unknown. METHODS: In this systematic review and meta-analysis, we searched six bibliographic databases (MEDLINE, Embase, PsycINFO, CINAHL, Web of Science, and Scopus) and grey literature published between Jan 1, 1980, and Feb 21, 2024, to identify UK studies reporting epilepsy incidence or epilepsy-related adverse outcomes by socioeconomic factors (individual level or area level). We included longitudinal cohort studies, studies using routinely collected health-care data, cross-sectional studies, and matched cohort studies and excluded conference abstracts and studies not reporting empirical results in the review and meta-analysis. Multiple reviewers (KJB, EC, SER, WOP, and RHT) independently screened studies, KJB extracted data from included studies and a second reviewer (SM or EC) checked data extraction. We used Critical Appraisal Skills Programme checklists to assess quality. We used random-effects meta-analysis to pool incident rate ratios (IRRs) and synthesised results on adverse outcomes narratively. This study was registered on PROPSPERO (CRD42023394143). FINDINGS: We identified 2471 unique studies from database searches. We included 26 studies, ten of which reported epilepsy incidence and 16 reported epilepsy-related adverse outcomes according to socioeconomic factors. Misclassification, participation, and interpretive biases were identified as study quality limitations. Meta-analyses showed an association between socioeconomic deprivation and epilepsy incidence, with greater risks of epilepsy incidence in groups of high-deprivation (IRR 1·34 [95% CI 1·16-1·56]; I2=85%) and medium-deprivation (IRR 1·23 [95% CI 1·08-1·39]; I2=63%) compared with low-deprivation groups. This association persisted in the studies that only included children (high vs low: IRR 1·36 [95% CI 1·19-1·57]; I2=0%). Only two studies examined factors influencing epilepsy incidence. There is limited evidence regarding UK inequalities in adverse outcomes. INTERPRETATION: Socioeconomic inequalities in epilepsy incidence are evident in the UK. To develop an evidence-based public health response to epilepsy, further research is needed to understand the populations affected, factors determining the association, and the extent of inequalities in adverse outcomes. FUNDING: Epilepsy Research Institute UK.

Weight change associated with antiepileptic drugs.

AIM: To investigate antiepileptic drug (AED)-related weight changes in patients with epilepsy through a retrospective observational study. METHOD: We analysed the anonymised electronic primary care records of 1.1 million adult patients in Wales. We included patients aged 18 years and over with a diagnosis of epilepsy, whose body weight had been measured up to 12 months before starting, and between 3 and 12 months after starting, one of five AEDs. We calculated the weight difference after starting the AED for each patient. RESULTS: 1423 patients were identified in total. The mean difference between body weight after and before starting each AED (together with 95% CI and p values for no difference) were: carbamazepine (CBZ) 0.43 (-0.19 to 1.05) p=0.17; lamotrigine (LTG) 0.31 (-0.38 to 1.00) p=0.38; levetiracetam (LEV) 1.00 (0.16 to 1.84) p=0.02; sodium valproate (VPA) 0.74 (0.10 to 1.38) p=0.02; topiramate (TPM) -2.30 (-4.27 to -0.33) p=0.02. CONCLUSIONS: LEV and VPA were associated with significant weight gain, TPM was associated with significant weight loss, and LTG and CBZ were not associated with significant weight change.

Development and validation of an automated basal cell carcinoma histopathology information extraction system using natural language processing.

Introduction: Routinely collected healthcare data are a powerful research resource, but often lack detailed disease-specific information that is collected in clinical free text such as histopathology reports. We aim to use natural Language Processing (NLP) techniques to extract detailed clinical and pathological information from histopathology reports to enrich routinely collected data. Methods: We used the general architecture for text engineering (GATE) framework to build an NLP information extraction system using rule-based techniques. During validation, we deployed our rule-based NLP pipeline on 200 previously unseen, de-identified and pseudonymised basal cell carcinoma (BCC) histopathological reports from Swansea Bay University Health Board, Wales, UK. The results of our algorithm were compared with gold standard human annotation by two independent and blinded expert clinicians involved in skin cancer care. Results: We identified 11,224 items of information with a mean precision, recall, and F1 score of 86.0% (95% CI: 75.1-96.9), 84.2% (95% CI: 72.8-96.1), and 84.5% (95% CI: 73.0-95.1), respectively. The difference between clinician annotator F1 scores was 7.9% in comparison with 15.5% between the NLP pipeline and the gold standard corpus. Cohen's Kappa score on annotated tokens was 0.85. Conclusion: Using an NLP rule-based approach for named entity recognition in BCC, we have been able to develop and validate a pipeline with a potential application in improving the quality of cancer registry data, supporting service planning, and enhancing the quality of routinely collected data for research.

Chronic inflammatory demyelinating polyneuropathy: a rare cause of falls.

This case of chronic inflammatory demyelinating polyneuropathy (CIDP) shows that a patient's condition can evolve from the point of admission, gradually manifesting its underlying cause. Our patient's initial presentation of backpain and lower limb weakness prompted investigations which ruled out compressive myelopathy and neuropathy. As upper limb weakness developed later, along with a more proximal and symmetrical pattern of lower limb weakness, the clinical picture suggested polyneuropathy. The diagnosis of CIDP became apparent only after numerous negative tests and nerve conduction studies which identified demyelination. Diagnosing CIDP enabled the commencement of definitive treatment which led to a good recovery.

Using natural language processing to extract structured epilepsy data from unstructured clinic letters: development and validation of the ExECT (extraction of epilepsy clinical text) system.

OBJECTIVE: Routinely collected healthcare data are a powerful research resource but often lack detailed disease-specific information that is collected in clinical free text, for example, clinic letters. We aim to use natural language processing techniques to extract detailed clinical information from epilepsy clinic letters to enrich routinely collected data. DESIGN: We used the general architecture for text engineering (GATE) framework to build an information extraction system, ExECT (extraction of epilepsy clinical text), combining rule-based and statistical techniques. We extracted nine categories of epilepsy information in addition to clinic date and date of birth across 200 clinic letters. We compared the results of our algorithm with a manual review of the letters by an epilepsy clinician. SETTING: De-identified and pseudonymised epilepsy clinic letters from a Health Board serving half a million residents in Wales, UK. RESULTS: We identified 1925 items of information with overall precision, recall and F1 score of 91.4%, 81.4% and 86.1%, respectively. Precision and recall for epilepsy-specific categories were: epilepsy diagnosis (88.1%, 89.0%), epilepsy type (89.8%, 79.8%), focal seizures (96.2%, 69.7%), generalised seizures (88.8%, 52.3%), seizure frequency (86.3%-53.6%), medication (96.1%, 94.0%), CT (55.6%, 58.8%), MRI (82.4%, 68.8%) and electroencephalogram (81.5%, 75.3%). CONCLUSIONS: We have built an automated clinical text extraction system that can accurately extract epilepsy information from free text in clinic letters. This can enhance routinely collected data for research in the UK. The information extracted with ExECT such as epilepsy type, seizure frequency and neurological investigations are often missing from routinely collected data. We propose that our algorithm can bridge this data gap enabling further epilepsy research opportunities. While many of the rules in our pipeline were tailored to extract epilepsy specific information, our methods can be applied to other diseases and also can be used in clinical practice to record patient information in a structured manner.

Treatment of resistant epilepsy.

Treatment resistance affects around 20% of people with epilepsy and carries a significant comorbidity. It is important to ensure that the diagnosis of epilepsy is secure and the underlying cause of the epilepsy is investigated thoroughly. Management involves early referral for epilepsy surgery when suitable, optimisation of pharmacological treatment, and consideration of comorbidities such as depression.